Commit Graph

232 Commits

Author SHA1 Message Date
Yorick Peterse 97d8450cba Removed the `regenerate` task. 2014-04-19 00:59:09 +02:00
Yorick Peterse 6f1ce17b31 Benchmark for lexer lines/second.
This benchmark uses a fixture file that is automatically downloaded.
2014-04-17 20:06:24 +02:00
Yorick Peterse 54e6650338 Don't use define_method in the lexer.
Profiling showed that calls to methods defined using `define_method` are
really, really slow. Before this commit the lexer would process 3000-4000
lines per second. With this commit that has been increased to around 10 000
lines per second.

Thanks to @headius for mentioning the (potential) overhead of define_method.
2014-04-17 19:08:26 +02:00
Yorick Peterse d9fa4b7c45 Lex input as a sequence of bytes.
Instead of lexing the input as a raw String or as a set of codepoints it's
treated as a sequence of bytes. This removes the need of String#[] (replaced by
String#byteslice) which in turn reduces the amount of memory needed and speeds
up the lexing time.

Thanks to @headius and @apeiros for suggesting this and rubber ducking along!
2014-04-17 17:45:05 +02:00
Yorick Peterse 70516b7447 Yield tokens in the lexer and parser.
After some digging I found out that Racc has a method called `yyparse`. Using
this method (and a custom callback method) you can `yield` tokens as a form of
input. This makes it a lot easier to feed tokens as a stream from the lexer.

Sadly the current performance of the lexer is still total garbage. Most of the
memory usage also comes from using String#unpack, especially on large XML
inputs (e.g. 100 MB of XML). It looks like the resulting memory usage is about
10x the input size.

One option might be some kind of wrapper around String. This wrapper would have
a sliding window of, say, 1024 bytes. When you create it the first 1024 bytes
of the input would be unpacked. When seeking through the input this window
would move forward.

In theory this means that you'd only end up with having only 1024 Fixnum
instances around at any given time instead of "a very big number". I have to
test how efficient this is in practise.
2014-04-17 00:39:41 +02:00
Yorick Peterse 144c95cbb4 Replaced the HRS fixture with one from Gist.
The HRS output is invalid, which Oga can not handle at this time.
2014-04-10 21:31:01 +02:00
Yorick Peterse 25edd2de00 Use a Set for storing void element names. 2014-04-10 12:28:47 +02:00
Yorick Peterse b96f7c4852 Lex attributes with namespaces.
These are lexed as just the name instead of two separate tokens.
2014-04-10 11:01:49 +02:00
Yorick Peterse c974b96b88 Truncate lines in parser errors.
The offending lines of code displayed in the error message are truncated to 80
characters. This should make reading the error messages less of a pain when
dealing with very long lines of HTML/XML.
2014-04-10 10:08:51 +02:00
Yorick Peterse 292a98d7f6 Basic benchmarks for the Parser class. 2014-04-10 10:05:04 +02:00
Yorick Peterse 8ca7781842 Updated the lexer benchmarks.
These had to be updated for the API changes of Oga::XML::Lexer.
2014-04-10 10:01:11 +02:00
Yorick Peterse 8237d5791d Stream tokens when lexing.
Instead of returning the tokens as a whole they are now streamed using
XML::Lexer#advance. This method returns the next token upon every call. It uses
a small buffer in case a particular block of text results in multiple tokens.
2014-04-09 22:08:13 +02:00
Yorick Peterse e9bb97d261 First steps towards making the lexer stream tokens 2014-04-09 19:32:06 +02:00
Yorick Peterse 10d0ec1573 Specs for parsing various empty nodes. 2014-04-07 21:33:23 +02:00
Yorick Peterse cb74c7edf9 Specs for XML parser errors. 2014-04-07 21:31:36 +02:00
Yorick Peterse 915d3ee505 Expanded tests for XML::Document#inspect. 2014-04-07 20:11:12 +02:00
Yorick Peterse e9412c9c4e Tests for various inspect methods. 2014-04-07 09:58:31 +02:00
Yorick Peterse 54ef125637 Basic docs for everything under Oga::XML. 2014-04-04 17:48:36 +02:00
Yorick Peterse 13a9228563 Properly indent doctype/XML decl inspect values. 2014-04-04 11:13:39 +02:00
Yorick Peterse 37a12722cb Rough setup for a custom #inspect format.
This format is a lot more readable than the default Ruby #inspect format
(mostly due to not including previous/next/parent nodes).
2014-04-04 00:41:29 +02:00
Yorick Peterse a2c525dd7c Insert newlines after XML dec/doctypes. 2014-04-03 23:04:21 +02:00
Yorick Peterse 230fafa2d3 Document should not inherit from Node.
A document is not an XML node on itself. If logic has to be shared between the
Document and the Node class I'll resort to using mixins for this.
2014-04-03 22:45:40 +02:00
Yorick Peterse c077988dd6 Tree building of doctypes. 2014-04-03 22:44:00 +02:00
Yorick Peterse 81b1155af3 Lex/parse doctype names separately. 2014-04-03 21:59:57 +02:00
Yorick Peterse 8185656c1e Fixed typ. 2014-04-03 21:41:31 +02:00
Yorick Peterse 6cf906e500 Lexer tests for single quoted attributes. 2014-04-03 18:50:07 +02:00
Yorick Peterse 30c01a5aee Tests for XML::TreeBuilder#handler_missing. 2014-04-03 09:43:30 +02:00
Yorick Peterse 0f129ceac9 Tests for XML::TreeBuilder#on_comment. 2014-04-03 09:38:18 +02:00
Yorick Peterse bdb76cefc5 Dedicated handling of XML declaration nodes. 2014-04-02 22:30:45 +02:00
Yorick Peterse d6c0a1f3f3 Lex/parser XML declaration attributes. 2014-04-02 22:01:17 +02:00
Yorick Peterse fa2e71c790 Tests for TreeBuilder#on_document. 2014-03-28 18:52:08 +01:00
Yorick Peterse f99c13b516 Tests + docs for the TreeBuilder class. 2014-03-28 17:11:54 +01:00
Yorick Peterse 6d866523b8 Renamed XML::Builder to XML::TreeBuilder. 2014-03-28 16:37:37 +01:00
Yorick Peterse 331726b2ca Tests for the various XML node types. 2014-03-28 16:34:30 +01:00
Yorick Peterse c366a96ce8 Rake task for generating code coverage. 2014-03-28 16:33:47 +01:00
Yorick Peterse e141c084f9 Dedicated DOM builder class for CDATA tags. 2014-03-28 09:27:53 +01:00
Yorick Peterse 2b250bbf42 Rough DOM building setup. 2014-03-28 08:59:48 +01:00
Yorick Peterse 6ae52c1b12 Initial rough sketches for the DOM API. 2014-03-26 18:12:00 +01:00
Yorick Peterse 6c661f3ee9 Removed the donations section.
I gave this some thought and I've removed it for two reasons:

1. My Dogecoin Wallet takes *forever* to sync with the network (13 weeks
   behind) so I uninstalled it. I can't be bothered waiting forever for a
   gimmick.

2. I don't like asking for donations/money. I'd much rather have people send me
   an Email thanking me for my work than for them to donate money. The latter
   means much more to me.
2014-03-25 23:55:10 +01:00
Yorick Peterse 4a48647d1e Removed generated lexer/parser.
I am a dumbass.
2014-03-25 21:47:40 +01:00
Yorick Peterse fb626278a8 Re-wrapped comments in the XML lexer. 2014-03-25 10:12:39 +01:00
Yorick Peterse 8ebd72158c Renamed XML::Lexer#t to #emit(). 2014-03-25 09:42:52 +01:00
Yorick Peterse 79818eb349 Added a convenience class for parsing HTML.
This removes the need for users having to set the `:html` option themselves.
2014-03-25 09:40:24 +01:00
Yorick Peterse 58009614f6 Moved XML specs into spec/oga/xml. 2014-03-25 09:36:39 +01:00
Yorick Peterse 7c03de0e2f Renamed HTML_PARSER to PARSER_OUTPUT.
This keeps it consistent with the lexer.
2014-03-25 09:35:48 +01:00
Yorick Peterse eae13d21ed Namespaced the lexer/parser under Oga::XML.
With the upcoming XPath and CSS selector lexers/parsers it will be confusing to
keep these in the root namespace.
2014-03-25 09:34:38 +01:00
Yorick Peterse 2259061c89 Don't require the 2nd Lexer#add_token argument. 2014-03-24 21:35:47 +01:00
Yorick Peterse 641c54261e Simplified lexer output for comments. 2014-03-24 21:34:30 +01:00
Yorick Peterse eaf1669b07 Simplified lexer output for CDATA tags. 2014-03-24 21:33:05 +01:00
Yorick Peterse 470be5a839 Simplified the lexer output for doctypes. 2014-03-24 21:32:16 +01:00